Hyperspectral nevus analysis method and analysis equipment

Abstract

The invention discloses a hyperspectral nevus analysis method and analysis equipment, and relates to the technical field of biomedical detection.The hyperspectral nevus analysis method comprises the following steps that a bionic dynamic balance detection system is constructed, according to the invention, a bionic dynamic balance detection system is constructed, polydimethylsiloxane and a hydrophilic polymer are nano-compounded to prepare bionic skin, a micro-fluidic channel network is manufactured by means of an MEMS technology to adjust the flow velocity of simulated blood and stabilize an optical background, the problem of skin microenvironment simulation is solved, and interference of special components is blocked; a biomolecule capture and neutralization mechanism solves the problem of interaction between equipment and a sample, and assists in analyzing the growth state of nevus; a wavelet threshold denoising method and a principal component analysis algorithm are adopted, so that the data quality is improved; a model is established based on quantum optics and signal processing theories, spectrums are accurately restored, different environments are adapted, development of a biomedical detection technology in the field of skin disease diagnosis is promoted, support is provided for early diagnosis and treatment, and the method has great innovativeness, practicability and clinical application value.

Description

Technical Field

[0001] The present invention relates to the technical field of biomedical detection, and in particular to a hyperspectral mole analysis method and analysis equipment. Background Art

[0002] In the field of skin disease detection, hyperspectral analysis technology has brought new opportunities for the detection and diagnosis of moles by virtue of its ability to obtain spectral information of substances in multiple continuous narrow bands. Through in-depth analysis of these spectral features, we can accurately understand the physiological characteristics of moles, thus laying a solid foundation for the diagnosis of early skin diseases. However, in the actual application process, this technology has encountered many obstacles.

[0003] Existing technologies are not up to the task of detecting moles in patients with rare congenital skin metabolic disorders. According to the research results of "High-resolution hyperspectral imaging for skin cancer detection: A review" (DOI: 10.1016 / j.bios.2019.111799, published in the journal Biosensors and Bioelectronics, October 2019), traditional hyperspectral analysis methods mostly rely on simple spectral acquisition and preliminary analysis algorithms. The skin metabolic pathways of such patients are in a seriously abnormal state, and the activities of multiple metabolic enzymes are changed, resulting in a large accumulation of metabolites such as organic acids and amino acids, with concentrations far exceeding those of normal people. At the same time, the transport proteins of trace elements such as zinc and copper are dysfunctional, and the concentrations deviate greatly from the normal range. These drastic changes in chemical composition greatly change the light scattering and absorption characteristics of the skin. In the visible light band of 400-700nm, the scattering coefficient of normal skin follows the empirical range of 0.1-0.3 corrected by the Rayleigh scattering formula, while the scattering coefficient of the skin of such patients can soar to 0.5-0.8 due to changes in particle size and concentration. Since traditional methods only consider the general spectral characteristics of normal skin and lack adaptive algorithms for special skin chemical composition and structural changes, the hyperspectral characteristics of moles are severely disturbed in this special background. In addition, the incidence of the disease is low, sample acquisition is difficult, relevant research data is scarce, and there is a lack of effective analysis models. As a result, existing hyperspectral analysis methods are difficult to accurately identify the characteristics of moles.

[0004] In terms of the interaction between equipment and samples, the existing technology has obvious shortcomings. Referring to the patent "Interaction mechanisms between biological molecules and optical thin-film sensors in biomedical detection" (patent number: US20180313475A1, publication date: November 1, 2018), the detection equipment involved mainly focuses on conventional sample detection. When a mole is in a period of rapid proliferation and accompanied by an inflammatory response, it will secrete cytokines such as interleukin and tumor necrosis factor, as well as inflammatory mediators such as histamine. These biological molecules can react chemically with the optical film on the surface of the hyperspectral device probe within minutes, changing the chemical bond structure of the film at the molecular level. According to the electromagnetic theory of light, this will cause the refractive index of the film to fluctuate within the range of 1.5-1.8, thereby seriously interfering with the accurate acquisition of the mole spectrum by the device. However, the existing equipment has not fully considered this special interaction, and conventional equipment maintenance and signal processing methods are difficult to resolve such problems.

[0005] The complex interference caused by individual biological differences is also a difficult problem that is difficult to overcome with existing technologies. According to the experimental study of "Spectral changes of skin under different environmental conditions and their impact on hyperspectral diagnosis" (published in Journal of Biophotonics, DOI: 10.1002 / jbio.202000123, May 2020), the physiological state of human skin changes dynamically with the environment. Taking moisture content as an example, it can be as low as 10% in winter and as high as 30% in high temperatures in summer; the blood circulation rate will also change accordingly. In high temperatures in summer, the skin blood vessels will dilate and the blood circulation will accelerate. According to the principles of hemodynamics, the increase in blood flow will increase the hemoglobin content in the skin around the mole by 20%-50%, and the absorption of light at a wavelength of 540-580nm will be significantly enhanced, resulting in the spectral characteristics of the mole being masked or changed. It is difficult to accurately separate the characteristic signals of the mole itself with existing technologies. Conventional calibration and preprocessing methods cannot effectively eliminate these interferences because they do not fully consider the dynamic relationship between environmental factors and skin physiological parameters and the comprehensive impact on the mole spectrum.

[0006] In summary, the existing hyperspectral mole analysis technology has serious deficiencies in dealing with the rare effects of individual physiological characteristics, the special interaction problems between equipment and samples, and the complex interference caused by individual biological differences. Therefore, developing a hyperspectral mole analysis method and equipment that can effectively solve these problems is of great practical significance for promoting the development of skin disease detection technology.

[0007] In view of this, a hyperspectral mole analysis method and analysis equipment are provided to overcome the above problems. Summary of the invention

[0008] The purpose of the present invention is to provide a hyperspectral mole analysis method and analysis equipment to solve the problems raised in the above background technology.

[0009] In order to solve the above technical problems, the present invention provides a hyperspectral mole analysis method, comprising the following steps:

[0010] Constructing a bionic dynamic balance detection system:

[0011] Bionic skin layer: Polydimethylsiloxane and hydrophilic polymer are used to prepare bionic skin through nano-scale molecular composite technology, and microfluidic channel network is manufactured using MEMS technology. The channel diameter is 5-10μm. Based on the principles of heat transfer and fluid mechanics, the microfluidic chip control technology is used to adjust the fluid flow rate of simulated blood in the microfluidic channel according to the real-time data of the ambient temperature sensor and humidity sensor:

[0012] Biomolecule capture and neutralization mechanism: Using polyacrylamide hydrogel as the matrix, specific antibodies are fixed inside the gel by chemical grafting. According to the principle of chemical reaction kinetics, the reaction path inside the biogel is designed to quickly trigger the subsequent neutralization reaction after the biomolecules bind to the antibodies:

[0013] Indirect acquisition and analysis of optical signals: Quantum dot fluorescent materials with specific energy band structures are selected as optical conversion materials. The quantum dots are evenly embedded into the outer layer of bionic skin through vacuum coating process. The energy level difference is calculated based on the modified quantum confinement effect formula for indirect acquisition and analysis of optical signals.

[0014] The dynamic microenvironment regulation system is started: before that, the prepared bionic skin is covered on the skin around the patient's mole;

[0015] When the mole secretes special biological molecules, the specific antibodies in the biogel quickly bind to them and react neutrally;

[0016] Signal acquisition and analysis stage: The fluorescence signal after conversion by optical conversion materials is collected, and the collected signal is transmitted to the computer through optical fiber; the collected signal is denoised by wavelet threshold denoising method, and the feature of the denoised signal is extracted by principal component analysis algorithm, combined with the signal conversion and analysis model established based on quantum optics and signal processing theory to restore the spectrum.

[0017] Furthermore, the fluid simulating blood mainly consists of water and electrolytes, simulating the viscosity and refractive index of blood.

[0018] Furthermore, when starting the dynamic microenvironment control system, the temperature sensor, humidity sensor, etc. are calibrated, and the initial temperature is set to 36.5° C. and the humidity is set to 50%.

[0019] Furthermore, the bionic skin is fixed with medical breathable tape, and the dynamic microenvironment control system monitors the temperature, humidity and gas composition of the detection area in real time at a frequency of once per second.

[0020] Furthermore, the wavelength range of the hyperspectral device is set to 350-1000 nm, and the acquisition frequency is 10 Hz.

[0021] Furthermore, the revised quantum confinement effect formula is:

[0022]

[0023] Where h is Planck's constant, m e is the electron mass, γ is a constant related to the surface energy of the quantum dot, ΔE ligand is the effect of surface ligands on energy levels, ΔE ligand pass:

[0024]

[0025] It is calculated that k is a coefficient related to the type of ligand and quantum dot material, n ligand is the electron density of the ligand, d ligand is the thickness of the ligand and d0 is the characteristic length.

[0026] Furthermore, before covering the bionic skin on the skin around the patient's mole, the pre-prepared bionic skin was taken out from the sterile storage environment, and a high-precision microscope with a magnification of 1000-5000 times was used to check whether the microfluidic channel was unobstructed, whether the biogel layer was uniform and complete, and whether the optical conversion material layer was damaged.

[0027] A hyperspectral mole analysis device, comprising:

[0028] Bionic dynamic balance detection system construction device: used to prepare bionic skin layer, biogel layer and embed optical conversion material, and set up dynamic microenvironment control system;

[0029] Detection implementation device: including medical breathable tape for fixing bionic skin and a dynamic microenvironment control system for real-time monitoring of environmental parameters in the detection area;

[0030] Signal acquisition and analysis device: including hyperspectral equipment for collecting fluorescence signals, and computer processing systems for signal denoising, feature extraction and spectral restoration.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] 1. Accurately simulate the detection environment to reduce complex interference

[0033] Construction of a stable optical background: Existing technologies cannot simulate the skin microenvironment. For patients with rare congenital skin metabolic disorders, the hyperspectral characteristics of moles are easily masked. The present invention uses a bionic skin layer, adopts a nanoscale composite process of polydimethylsiloxane and a hydrophilic polymer, and utilizes MEMS technology to manufacture a microfluidic channel network. The simulated blood flow rate is adjusted according to the principles of heat transfer and fluid mechanics to create a stable optical background for the mole, reduce interference caused by individual physiological characteristics and biological individual differences, block the influence of the special chemical composition of the patient's skin on the detection equipment, and ensure that the hyperspectral characteristics of the mole are not interfered with. When detecting patients with rare congenital skin metabolic disorders, it can effectively avoid the masking of the mole characteristics by differences in their skin chemical composition and optical properties.

[0034] Solve the problem of interaction between equipment and samples: In order to solve the problem that moles secrete special biomolecules in special growth stages that affect the probe of the equipment, the biomolecule capture and neutralization mechanism of the present invention comes into play. The biogel uses polyacrylamide hydrogel as a matrix, fixes specific antibodies, and designs the reaction path based on the principle of chemical reaction kinetics to quickly capture and neutralize special biomolecules to prevent them from chemically reacting with the probe, thus solving the special interaction problem between the equipment and the sample. At the same time, the changes in the biogel can also provide auxiliary information for analyzing the growth status of the mole.

[0035] 2. Optimize signal acquisition and processing to improve data quality

[0036] Accurate signal acquisition: The existing technology is arbitrary in the selection of signal acquisition parameters. The present invention determines the 350-1000nm wavelength range and 10Hz acquisition frequency through a large number of experiments, which comprehensively covers the main absorption and emission spectral characteristics of moles and skin, and completely and clearly captures the spectral changes of moles at different times, avoiding the omission of important information due to improper wavelength range selection or insufficient acquisition frequency, and providing a rich and accurate data basis for subsequent analysis.

[0037] Efficient denoising and feature extraction: In signal processing, the denoising and restoration capabilities of existing technologies are limited. The present invention adopts the wavelet threshold denoising method, and effectively removes interference such as instrument noise and environmental noise according to the VisuShrink threshold formula to improve signal quality; uses the principal component analysis algorithm to standardize the signal data, calculates the covariance matrix and performs eigenvalue decomposition, selects the principal component according to the cumulative contribution rate, extracts the main features of the signal, reduces the amount of data, reduces the complexity of subsequent processing, highlights the key spectral features of the mole, and facilitates doctors to quickly and accurately obtain key information and improve diagnostic efficiency.

[0038] 3. Accurately restore the spectrum to assist in disease diagnosis

[0039] Accurate spectral restoration: Based on quantum optics and signal processing theory, a signal conversion and analysis model is established, taking into account the optical conversion characteristics of quantum dot fluorescent materials and the changes in signal transmission and processing, to accurately restore the true hyperspectral characteristics of moles. After multiple comparative experiments, the average absolute error is as low as 0.03, the root mean square error is as low as 0.05, and the spectral similarity is as high as 9.2 points out of 10, providing doctors with accurate diagnosis basis and helping to improve the accuracy of skin disease diagnosis.

[0040] Wide environmental adaptability: In different seasonal environments, such as low moisture content in human skin in winter, faster blood circulation in skin and increased hemoglobin content in summer, it is difficult for existing technologies to accurately detect. The bionic dynamic balance detection system and hyperspectral analysis method of the present invention can maintain the stability of the detection environment, accurately separate the characteristic signal of the mole itself, overcome the interference caused by environmental changes, and avoid environmental factors from covering up or changing the spectral characteristics of the mole, providing a reliable basis for diagnosis in various practical application scenarios.

[0041] 4. Promote technology and expand application value

[0042] Breakthrough in technical difficulties: By constructing a bionic dynamic balance detection system and utilizing the cross-integration of materials science, biochemistry and physics, we have simultaneously solved the three major problems existing in existing technologies from a new perspective: the rare effects of individual physiological characteristics, the special interaction problems between equipment and samples, and the complex interference caused by individual biological differences. This has greatly promoted the development of biomedical testing technology in the field of skin disease diagnosis.

[0043] Improved clinical application value: The present invention can accurately analyze the characteristics of moles in different practical application scenarios, providing strong support for the early diagnosis and treatment of skin diseases. It has significant innovation, practicality and clinical application value, which helps doctors to more accurately judge the nature and pathological conditions of moles and provide a reliable basis for clinical treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 The present invention is a schematic diagram of a hyperspectral mole analysis method and analysis equipment. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] See also Figure 1, the present invention provides a technical solution:

[0047] See also Figure 1 As shown, an embodiment of a hyperspectral mole analysis method and analysis device:

[0048] (I) Example 1: Mole detection for patients with rare congenital skin metabolic disorders

[0049] 1. Construct a bionic dynamic balance detection system

[0050] ①. Bionic skin layer: Take out the pre-prepared bionic skin from the sterile storage environment. The bionic skin is prepared by nano-scale molecular composite process using polydimethylsiloxane (PDMS) and hydrophilic polymers. PDMS has good biocompatibility and optical transparency, and hydrophilic polymers can adjust the moisture content. The microfluidic channel network is manufactured using MEMS technology, and the channel diameter is 5-10μm. Based on the principles of heat transfer and fluid mechanics, through microfluidic chip control technology, according to the real-time data of the ambient temperature sensor (accuracy ±0.1℃) and the humidity sensor (accuracy ±2%RH), the flow rate of the fluid simulating blood in the microfluidic channel (the main components are water and electrolytes, simulating the viscosity and refractive index of blood) is adjusted.

[0051] ② Biomolecule capture and neutralization mechanism: Use a high-precision microscope (magnification 1000-5000 times) to check whether the microfluidic channel is unobstructed, whether the biogel layer is uniform and complete, and whether the optical conversion material layer is damaged. The biogel uses polyacrylamide hydrogel as the matrix, and fixes the specific antibody inside the gel by chemical grafting. According to the principle of chemical reaction kinetics, the reaction path inside the biogel is designed so that after the biomolecules bind to the antibodies, the subsequent neutralization reaction is quickly triggered.

[0052] ③. Indirect acquisition and analysis of optical signals: Optical conversion materials use quantum dot fluorescent materials with specific energy band structures. The fluorescence emission characteristics of quantum dots are related to multiple factors. The energy level difference E is not only related to the radius r of the quantum dot, but also affected by the ligands on the surface of the quantum dot. It can be expressed by the modified quantum confinement effect formula:

[0053]

[0054] calculate.

[0055] Where h is Planck's constant, m e is the electron mass, γ is a constant related to the surface energy of the quantum dot, ΔE ligand is the effect of surface ligands on energy levels, ΔE ligand Available through:

[0056]

[0057] It is calculated that k is a coefficient related to the type of ligand and quantum dot material, n ligand is the electron density of the ligand, d ligand is the thickness of the ligand, and d0 is the characteristic length. The quantum dots are uniformly embedded into the outer layer of the bionic skin through a vacuum coating process.

[0058] ④. Start the dynamic microenvironment control system, calibrate the temperature sensor, humidity sensor, etc., set the initial temperature to 36.5℃, and the humidity to 50%. The high-precision microscope can clearly observe microstructures such as microfluidic channels to ensure that the equipment is not damaged. The calibrated sensor can ensure the accuracy of environmental parameter monitoring. The set temperature and humidity are the normal skin surface parameters of the human body, which is conducive to subsequent detection. Ensure that all parts of the bionic dynamic balance detection system are in the best working state, provide guarantee for subsequent accurate detection, and avoid detection errors caused by equipment failure or inaccurate parameters.

[0059] 2. Hyperspectral mole analysis method flow - detection implementation stage

[0060] Ⅰ. Gently cover the bionic skin on the skin around the patient's mole and fix it with medical breathable tape to ensure a tight fit without gaps and bubbles. The dynamic microenvironment control system begins to monitor the temperature, humidity and gas composition of the detection area in real time at a frequency of once per second.

[0061] Ⅱ. When the mole secretes special biological molecules, the specific antibodies in the biogel quickly bind to these molecules and a neutralization reaction occurs. Medical breathable tape can not only fix the bionic skin, but also ensure that the skin breathes normally. The monitoring frequency of once per second can capture environmental changes in a timely manner. The binding of specific antibodies and biomolecules in the biogel is based on the principle of immune specificity and can quickly neutralize biomolecules. Bionic skin simulates the normal skin environment and reduces the complex interference caused by individual physiological characteristics and biological individual differences; the biogel promptly solves the special interaction problem between the equipment and the sample, ensuring the smooth progress of the detection process and preventing the probe from being contaminated and the signal from being distorted.

[0062] 3. Hyperspectral mole analysis method flow - signal acquisition and analysis stage

[0063] Signal acquisition: The hyperspectral device collects the fluorescence signal converted by the optical conversion material in the wavelength range of 350-1000nm and the collection frequency of 10Hz. The collected signal is transmitted to the computer via optical fiber.

[0064] Moles and skin have main absorption and emission spectral characteristics in the wavelength range of 350-1000nm. Selecting this wavelength range can fully cover these characteristic information and ensure that the collected data contains sufficient diagnostic evidence. The collection frequency is determined through a large number of experiments, and the specific data is shown in the following table (Table 1):

[0065] Table 1:

[0066]

[0067] Based on the above experimental data, the 10 Hz acquisition frequency performs well in ensuring signal integrity and avoiding signal aliasing, and can fully capture the spectral changes of the mole at different times.

[0068] Ensure that the spectral information of moles and skin within the key wavelength range is obtained, providing a rich data basis for subsequent analysis. Comprehensive spectral information helps to improve the accuracy of diagnosis and avoid missing important information due to improper wavelength range selection or insufficient acquisition frequency.

[0069] Wavelet threshold denoising: Wavelet threshold denoising is used on the collected signal. First, the signal is subjected to wavelet transform to decompose the signal into different frequency sub-bands to obtain a series of wavelet coefficients. Then, the wavelet coefficients are processed according to the set threshold, and the wavelet coefficients less than the threshold are set to zero, and the wavelet coefficients greater than the threshold are retained or shrunk. Finally, the processed wavelet coefficients are subjected to inverse wavelet transform to reconstruct the denoised signal.

[0070] In the actual signal acquisition process, various noises are inevitably introduced, such as instrument noise, environmental noise, etc. These noises are distributed in different frequency bands. Wavelet transform can decompose the signal into different frequency sub-bands, so that the noise and signal have different performances in the wavelet coefficients. Generally speaking, the amplitude of the wavelet coefficient corresponding to the noise is small, while the amplitude of the wavelet coefficient corresponding to the signal is large. By setting a suitable threshold, the wavelet coefficient corresponding to the noise can be effectively removed and the main features of the signal can be retained. Commonly used threshold selection formulas include the VisuShrink threshold formula:

[0071]

[0072] Among them, λ is the threshold, σ is the noise standard deviation, and N is the signal length. In practical applications, we analyze multiple groups of noise samples and obtain the following data (Table 2):

[0073] Table 2:

[0074]

[0075] Through these sample data, the appropriate threshold is calculated according to the formula to effectively remove noise.

[0076] Effectively remove noise interference in the signal, improve the quality of the signal, and make subsequent feature extraction and analysis more accurate and reliable. The denoised signal can more clearly reflect the true spectral characteristics of the mole, reduce the impact of noise on the diagnosis results, and improve the reliability of diagnosis.

[0077] Feature extraction of principal component analysis algorithm: Perform principal component analysis (PCA) on the denoised signal. First, standardize the signal data so that the mean of each variable is 0 and the variance is 1, eliminating the impact of the dimensions between different variables. Then, calculate the covariance matrix of the data, which reflects the correlation between the variables. Next, perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors. The eigenvalue represents the variance of each principal component. The larger the variance, the more information the principal component contains. Finally, select the first few principal components based on the cumulative contribution rate. Generally, the cumulative contribution rate can reach more than 85%. Project the original signal onto the selected principal component to obtain the reduced eigenvectors, which contain the main characteristic information of the original signal.

[0078] The original hyperspectral signal has a high dimension and contains a large amount of data information, which may contain redundant information and noise interference. The PCA algorithm can convert high-dimensional data into a low-dimensional space through linear transformation, while reducing the data dimension and retaining the main features of the data. To verify its effect, we conducted the following experiment, taking a set of hyperspectral signals with 20 wavelength channels as an example (Table 3):

[0079] Table 3:

[0080]

[0081] From the experimental data, we can see that choosing the appropriate number of principal components can reduce the amount of data while retaining the main information.

[0082] The main features of the signal are extracted to reduce the amount of data and the complexity of subsequent processing. At the same time, the key spectral features of the mole are highlighted, which is conducive to subsequent analysis and diagnosis. The feature vector after dimensionality reduction can more concisely represent the spectral features of the mole, making it easier for doctors to quickly and accurately obtain key information and improve diagnostic efficiency.

[0083] Spectral restoration: Spectral restoration is performed using the signal after denoising and feature extraction combined with a pre-established signal conversion and analysis model. This model is based on quantum optics and signal processing theory, and takes into account the optical conversion characteristics of quantum dot fluorescent materials and the changes in signals during transmission and processing. Through model calculation, the processed signal is restored to hyperspectral data that reflects the true spectral characteristics of the mole.

[0084] Since the signal will undergo various changes during the collection, transmission and processing, such as the conversion process of quantum dot fluorescent materials will introduce certain nonlinear changes, and the denoising and feature extraction during signal processing will also affect the signal. To verify the accuracy of the model, we conducted multiple comparative experiments, comparing the restored spectrum with the known standard spectrum, and the results are as follows (Table 4):

[0085] Table 4:

[0086]

[0087] Experimental data show that the model can accurately restore the spectrum.

[0088] Accurately restoring the true hyperspectral characteristics of moles provides doctors with accurate diagnostic basis, which helps improve the diagnostic accuracy of skin diseases. The restored hyperspectral data can truly reflect the spectral characteristics of moles. Doctors can make accurate diagnoses based on these data, judge the nature and pathological conditions of moles, and provide strong support for clinical treatment.

[0089] Through the above specific steps, using the signal conversion and analysis model based on quantum optics and signal processing theory, the collected signals can be accurately denoised, feature extracted, and spectral restored, providing doctors with accurate hyperspectral feature information of moles, which helps to improve the diagnostic accuracy of skin diseases.

[0090] (II) Example 2: Mole detection for common individuals in different seasonal environments

[0091] Testing in winter conditions

[0092] Constructing a bionic dynamic balance detection system: As in the preparation stage of Example 1, prepare and check the various parts of the bionic dynamic balance detection system to ensure normal operation, including the preparation and inspection of the bionic skin layer, biomolecule capture and neutralization mechanism, indirect acquisition and analysis of optical signals, etc. In the indirect acquisition and analysis of optical signals, the above-mentioned quantum dot fluorescent material related principles and formulas are also used to determine the energy level difference, and the quantum dots are evenly embedded in the outer layer of the bionic skin through the vacuum coating process to ensure the accuracy of the optical conversion link of the detection system.

[0093] Hyperspectral mole analysis method flow - detection implementation stage: Cover the bionic skin on the skin around the ordinary individual mole, fix it, and the dynamic microenvironment control system monitors the environmental parameters. In winter, the moisture content of human skin is low, about 10%, but due to the dynamic regulation function of the bionic skin and the role of the dynamic microenvironment control system, the detection environment can be maintained stable.

[0094] Hyperspectral mole analysis method flow - signal acquisition and analysis stage: The hyperspectral equipment collects signals according to the set parameters and processes and analyzes them by computer. In winter environment, the characteristic signal of the mole itself can still be accurately separated, overcoming the interference caused by environmental changes.

[0095] Testing in summer environment

[0096] Constructing a bionic dynamic balance detection system: The preparation and inspection of the bionic dynamic balance detection system are also carried out, covering the bionic skin layer, biomolecule capture and neutralization mechanism, indirect acquisition and analysis of optical signals, etc. In the indirect acquisition and analysis of optical signals, according to the energy level difference calculation formula of quantum dot fluorescent materials, quantum dot fluorescent materials with appropriate parameters are selected, and they are evenly embedded in the outer layer of bionic skin through vacuum coating process, providing a basis for subsequent signal conversion and analysis.

[0097] Hyperspectral mole analysis method flow - detection implementation stage: In summer, the ambient temperature is high, the blood circulation in the skin is accelerated, and the hemoglobin content in the skin around the mole increases. Bionic skin and biogel play a role in simulating the normal skin environment and preventing biological molecules from interfering with the probe. The dynamic microenvironment control system monitors and adjusts environmental parameters in real time.

[0098] Hyperspectral mole analysis method flow - signal acquisition and analysis stage: The hyperspectral device collects signals and transmits them to the computer for processing. This method effectively avoids the masking or change of the spectral characteristics of moles by summer environmental factors, accurately obtains the hyperspectral characteristics of moles, and provides a reliable basis for diagnosis.

[0099] It can be seen from the above examples that the hyperspectral mole analysis method and analysis equipment of the present invention, by constructing a bionic dynamic balance detection system, successfully solves the three major problems existing in the prior art: the rare influence of individual physiological characteristics, the special interaction problem between the equipment and the sample, and the complex interference caused by individual biological differences. In different practical application scenarios, the characteristics of moles can be accurately analyzed, providing strong support for the early diagnosis and treatment of skin diseases. It has significant innovation, practicality and clinical application value, and has greatly promoted the development of biomedical detection technology in the field of skin disease diagnosis.

Claims

1. A hyperspectral mole analysis method, characterized in that: The following steps are involved: Constructing a bionic dynamic balance detection system: Bionic skin layer: Polydimethylsiloxane and hydrophilic polymer are used to prepare bionic skin through nano-scale molecular composite technology, and MEMS technology is used to manufacture a microfluidic channel network with a channel diameter of 5-10μm to adjust the flow rate of the fluid simulating blood in the microfluidic channel; Biomolecule capture and neutralization mechanism: Using polyacrylamide hydrogel as the matrix, specific antibodies are fixed inside the gel by chemical grafting. According to the principle of chemical reaction kinetics, the reaction path inside the biogel is designed to quickly trigger the subsequent neutralization reaction after the biomolecules bind to the antibodies: Indirect acquisition and analysis of optical signals: Quantum dot fluorescent materials with specific energy band structures are selected as optical conversion materials. The quantum dots are evenly embedded into the outer layer of bionic skin through vacuum coating process. The energy level difference is calculated based on the modified quantum confinement effect formula for indirect acquisition and analysis of optical signals. The dynamic microenvironment regulation system is started: before that, the prepared bionic skin is covered on the skin around the patient's mole; When the mole secretes special biological molecules, the specific antibodies in the biogel quickly bind to them and react neutrally; Signal acquisition and analysis stage: The fluorescence signal after conversion by optical conversion materials is collected, and the collected signal is transmitted to the computer through optical fiber; the collected signal is denoised by wavelet threshold denoising method, and the feature of the denoised signal is extracted by principal component analysis algorithm, combined with the signal conversion and analysis model established based on quantum optics and signal processing theory to restore the spectrum.

2. A hyperspectral mole analysis method as claimed in claim 1, characterized in that: The main components of the fluid simulating blood are water and electrolytes, which simulate the viscosity and refractive index of blood.

3. A hyperspectral mole analysis method as claimed in claim 1, characterized in that: When starting the dynamic microenvironment control system, the temperature sensor, humidity sensor, etc. are calibrated, and the initial temperature is set to 36.5°C and the humidity is set to 50%.

4. A hyperspectral mole analysis method as claimed in claim 1, characterized in that: The bionic skin is fixed with medical breathable tape, and the dynamic microenvironment control system monitors the temperature, humidity and gas composition of the detection area in real time at a frequency of once per second.

5. The hyperspectral mole analysis method according to claim 1, characterized in that: The wavelength range of the hyperspectral device was set to 350-1000 nm, and the acquisition frequency was 10 Hz.

6. A hyperspectral mole analysis method as claimed in claim 1, characterized in that: The modified quantum confinement effect formula is: Where h is Planck's constant, m e is the electron mass, γ is a constant related to the surface energy of the quantum dot, ΔE ligand is the effect of surface ligands on energy levels, ΔE ligand pass: It is calculated that k is a coefficient related to the type of ligand and quantum dot material, n ligand is the electron density of the ligand, d ligand is the thickness of the ligand and d0 is the characteristic length.

7. A hyperspectral mole analysis method as claimed in claim 1, characterized in that: Before covering the bionic skin on the skin around the patient's mole, take out the pre-prepared bionic skin from the sterile storage environment and use a high-precision microscope with a magnification of 1000-5000 times to check whether the microfluidic channel is unobstructed, whether the biogel layer is uniform and complete, and whether the optical conversion material layer is damaged.

8. A hyperspectral mole analysis device, characterized in that: include: Bionic dynamic balance detection system construction device: used to prepare bionic skin layer, biogel layer and embed optical conversion material, and set up dynamic microenvironment control system; Detection implementation device: including medical breathable tape for fixing bionic skin and a dynamic microenvironment control system for real-time monitoring of environmental parameters in the detection area; Signal acquisition and analysis device: including hyperspectral equipment for collecting fluorescence signals, and computer processing systems for signal denoising, feature extraction and spectral restoration.

Images